CodeRabbit helps teams review, understand, govern, and trust agent-generated code: https://nitter.cf/t.co/GxSOapi2As!
San Francisco, CA
Joined April 2023
- Tweets6.9K
- Following108
- Followers48.7K
- Likes13.9K
Pinned Tweet
New in CodeRabbit Secure:
Deep Scan now follows a vulnerability across your linked repos, then replays the exploit in a throwaway sandbox to prove it's real before it flags anything.
Hendrik walks through this and everything else in the new update:
CodeRabbit Secure ran 13,000+ deep scans over the last 10 days and turned up 43,000+ high or critical vulnerabilities.
Nearly 10,000 of those are already fixed, which is the bit that matters.
Run your deep scan today, and fix hidden vulnerabilities in your code base!
CodeRabbit retweeted
We put GPT-6 Cyber inside CodeRabbit’s security scan pipeline.
22 of 33 known vulnerabilities caught vs 19 for our production ensemble, and 66% more high and medium severity findings.
Full eval below.
Introducing GPT‑6 Cyber: a major advance in security research that pairs stronger capabilities with better model alignment.
The model powered the Linux kernel research we announced yesterday, and security teams at companies like Mozilla, Palo Alto Networks, Calif, and Cisco—alongside our own at OpenAI—are already putting it to work.
GPT-6 Cyber is available to select Daybreak Red customers and we are expanding access over time.
openai.com/index/introducing…
We put GPT-6 Cyber inside CodeRabbit’s security scan pipeline.
22 of 33 known vulnerabilities caught vs 19 for our production ensemble, and 66% more high and medium severity findings.
Full eval below.
Introducing GPT‑6 Cyber: a major advance in security research that pairs stronger capabilities with better model alignment.
The model powered the Linux kernel research we announced yesterday, and security teams at companies like Mozilla, Palo Alto Networks, Calif, and Cisco—alongside our own at OpenAI—are already putting it to work.
GPT-6 Cyber is available to select Daybreak Red customers and we are expanding access over time.
openai.com/index/introducing…
"Just because you can, doesn't mean you should."
AI makes it easier to ship more, but Hassan El Mghari makes the case for shipping less: focus on one thing, make it work well, and resist adding features just because you can.
Watch our 29 minute conversation on Rabbits on the Run here!
Featuring @nutlope @JuanPa and a guest appearance by @theo!
CodeRabbit retweeted
Routine PR chores in @coderabbitai Triage can now run themselves
Rules:
+ merge low-risk changes once checks and reviews pass
+ let the Coding Agent auto-fix failing PRs
+ close stale ones after a 24h notice
+ ping reviewers in Slack
run it on your repos!
Building your AWS re:Invent schedule?
Come by CodeRabbit at booth #528 for a live demo and a conversation about your team’s code.
📆 Nov 30 – Dec 4
🌍 Las Vegas
coderabbit.ai/events/aws-rei…
CodeRabbit retweeted
From the @coderabbitai changelog
The Deep Scan findings API is now on all plans. No Enterprise plan, Security add-on or trial
All you need is a non-agentic API key with security:read
GET /v1/security/scans/code returns findings as JSON, CSV or SARIF
A 403 doesn’t mean the bug is fixed, it might mean your request never reached the code you were testing.
Runtime Verification in CodeRabbit Secure shows whether the request actually hit the target, right next to the verdict.
How we built it 👇
CodeRabbit retweeted
What else does your PR touch?
@coderabbitai Secure Blast Radius maps it in Change Stack
upstream, downstream and tests, every edge backed by code evidence + security findings light up the paths they touch
(in open beta)
